TW201805750A - Mobile device for correcting surrounding information autonomously and method for correcting - Google Patents

Abstract

The present invention relates to mobile device for correcting surrounding information autonomously, comprising: a data storage module, for storing the map of a moving area and corresponding map description file, the plurality of artificial mark to be set within the moving area; an image capture module, for capturing the forward images autonomously; a artificial mark identification module for identifying image of artificial markers, to achieve positioning; a data recording module, recording the road condition changeable informations and moving informations; a data analysis module to analyze a optimal moving information according to the moving informations; a data correction module, for correcting the maps and corresponding map description file according to road conditions changeable informations and the optimal moving information. The present invention further relates to a method for correcting surrounding information autonomously.

Description

Autonomous mobile equipment and method for automatically correcting environmental information

The invention relates to an autonomous mobile device and method for automatically correcting environmental information.

Real-time positioning and map construction (SLAM) is the most commonly used technology in the positioning process of autonomous mobile devices such as robots. The SLAM technology refers to the robot's positioning from unknown locations and locations through repeated observation of map features during the movement process. Its own position and attitude, and then build the map incrementally according to its own, so as to achieve the purpose of simultaneous positioning and map construction. The SLAM technology mostly relies on sensors to provide additional information to achieve positioning, such as GPS, IMU, Odometry (wheel speedometer). When an autonomous mobile device equipped with universal wheels or omnidirectional wheels moves, the speedometer cannot As a reference for moving distance, and indoor environment is not suitable to use GPS.

Previously, people used hand-tagged computer vision to achieve positioning, which basically could be done without using the IMU. However, when the same autonomous mobile device is in different environmental conditions, the same motor output cannot reach the same moving distance. Although the method of building correction can be used to make the autonomous mobile device reach the destination, the movement of the autonomous mobile device will Looks awkward.

In view of this, it is indeed necessary to provide an autonomous mobile device and method for automatically correcting environmental information so that the autonomous mobile device can move smoothly.

An autonomous mobile device that automatically corrects environmental information includes: a data storage module;

First store a map of the area to be moved and the corresponding map description file. A plurality of manual marks are set in the area to be moved, and the autonomous mobile device moves between the artificial marks; an image acquisition module is used to photograph the autonomous movement An image in front of the device during the movement; an artificial mark recognition module for receiving an image signal output by the image acquisition module and identifying the artificial mark in the image, thereby realizing the positioning of the autonomous mobile device; A data recording module records information about changes in the pavement environment between the artificial marks and movement information of autonomous mobile devices moving between the artificial marks; a data analysis module, according to the movement recorded by the data recording module The information analyzes the optimal movement information for the autonomous mobile device to move between the artificial marks; a data correction module is used to receive the road environment change information recorded by the data recording module and the most analyzed by the data analysis module. Optimize the signal of mobile information to correct the map and map description file in the data storage module.

A method for automatically correcting environmental information includes the following steps: providing an autonomous mobile device for automatically correcting environmental information, which includes a data storage module for storing a map of a region to be moved and a corresponding map description file in advance, A plurality of artificial markers are set in the region to be moved, and autonomous mobile devices move between the artificial markers; an image acquisition module is used to capture an image of the front of the autonomous mobile device during the movement process; an artificial marker recognition module For receiving an image signal output by the image acquisition module and recognizing an artificial mark in the image, thereby realizing the positioning of the autonomous mobile device; a data recording module for recording a road surface environment between the artificial marks Change information of the mobile device and mobile information of the autonomous mobile device moving between the artificial marks; a data analysis module that analyzes the autonomous mobile device moving between the artificial marks based on the mobile information recorded by the data recording module; Optimal movement information; a data correction module for receiving information on road surface environmental changes recorded by the data recording module The signal of the optimal mobile information analyzed by the data analysis module, so as to modify the map and map description file in the data storage module; start the autonomous mobile device to move between two manual markers, and the image acquisition The module collects the image in front of its movement and transmits it to the artificial recognition module; the artificial mark recognition module recognizes the artificial mark in the image, and the autonomous mobile device determines its position relative to the artificial mark The data recording module simultaneously records the change information of the road surface environment between the manual marks and the mobile information of the autonomous mobile device moving between the manual marks, and the data analysis module judges that the manual Optimal movement information for moving between markers; the data correction module corrects a map and a map description file in the data storage module according to the road surface environment change information and optimal movement information.

Compared with the prior art, the present invention provides an autonomous mobile device and method for automatically correcting environmental information. By storing a map of a region to be moved and a corresponding map description file in the autonomous mobile device in advance, it records when the autonomous mobile device moves between manual markers. Pavement environment change information and mobile information, based on the mobile information analysis, to obtain the optimal mobile information for autonomous mobile devices to move between the manual markers, and based on this, continuously modify the maps and map description files stored in the autonomous mobile devices. Thus, the autonomous mobile device can move more smoothly in the area to be moved.

FIG. 1 is a module diagram of an autonomous mobile device that automatically corrects environmental information according to the present invention.

FIG. 2 is a flowchart of a method for automatically correcting environmental information according to the present invention.

In the following, the autonomous mobile device and method for automatically correcting environmental information provided by the present invention will be further described in detail with reference to the drawings and specific embodiments.

Please refer to FIG. 1. The present invention provides an autonomous mobile device that automatically corrects environmental information.

Including: a data storage module for storing a map of a region to be moved and a corresponding map description file in advance, a plurality of manual markers are set in the region to be moved, and an autonomous mobile device moves between the manual markers; An image acquisition module is used to capture the image of the front of the autonomous mobile device during the movement; an artificial mark recognition module is used to receive the image signal output by the image acquisition module and recognize the artificial mark in the image, Thus, the positioning of the autonomous mobile device is realized; a data recording module records the change information and movement information of the road surface environment when the autonomous mobile device moves between the artificial marks; a data analysis module according to the data recording module The mobile information recorded by the group analyzes the optimal mobile information of the autonomous mobile device moving between the manual markers; a data correction module for receiving information on the change of the road surface environment and the data analysis module recorded by the data recording module Analyze the signal of the optimal mobile information to correct the map and map description file in the data storage module.

The autonomous mobile device may be a device capable of autonomous movement, such as a robot, an unmanned van, or an unmanned aerial vehicle. The autonomous mobile device may be moved in a foot type, a wheel type, or the like.

The area to be moved may be some fixed workplaces, such as factories, restaurants, patrol stations, and the like. A plurality of artificial marks are set in the area to be moved, and each artificial mark corresponds to an ID. The ID may be information such as digits and characters, and represents the name of the artificial mark, such as a corner. The manual tag is any one or more of a Tag36h11 tag series, a Tag36h10 tag series, a Tag25h9 tag series, or a Tag16h5 tag series.

The data storage module is used to store a map of the area to be moved and a corresponding map description file in advance. The map of the area to be moved is stored in a well-designed markup language (XML) or other format file, where manual markup is defined; the map description file includes a description of the area near the manual markup in the map, and Mobile information that moves smoothly between manual markers, such as the names of manually marked places in the map, information such as the distance, direction, and required motor output for autonomous mobile devices to move from one manual marker to the next. The data storage module can store only the revised mobile information for the same road environment, and can also store the revised mobile information on the basis of retaining the original mobile information, such as walking from manual mark A to manual mark B, the stored movement information is “Go straight from the artificial mark A 5 times forward and then back up to reach the artificial mark B”, the next time you find “Go straight 4 times from the artificial mark A to reach the artificial mark B ", You can use the mobile information to overwrite the stored mobile information.

The image acquisition module includes a camera, which is disposed on the side of the autonomous mobile device facing the direction of movement and captures an image of an area that can be covered by its own field of vision so that the artificial marker can be captured. The image is transmitted to the manual identification module through a data line. The camera may be a general CCD or CMOS-based web camera.

The artificial mark recognition module receives the image captured by the image acquisition module, reads and recognizes the artificial mark therein, and thereby determines the relative position and angle of the autonomous mobile device with respect to the artificial mark, thereby achieving Positioning. When the pavement environment between the artificial markers changes, the autonomous mobile device moves from the artificial marker A to the artificial marker B according to the stored map description file. When it is found that it is located at the lower left of the artificial marker B, the autonomous mobile device can autonomously locate at this time. Give the corresponding processing command to reach the manual marker B directly below; when the road surface environment between the manual markers has not changed, the autonomous mobile device moves from the manual marker A to the manual marker B according to the stored map description file, and just arrived Directly below the manual mark, there is no need for autonomous correction.

The data recording module is used to record the change information of the pavement environment between the manual marks during the movement and the corresponding movement information to move between the manual marks. The change information of the pavement environment includes the increase or disappearance of obstacles and the change of the road friction coefficient. For example, some obstacles such as tables, chairs, or carpets, sand, stones, etc. are suddenly added to the map. When the pavement environment changes, the movement information specifically includes the number of displacement corrections and the corresponding displacement correction amount when the autonomous mobile device moves between manual markers. If there is an additional table, the data recording module records obstacles. Number of displacement corrections for the object and the autonomous mobile device to bypass the new obstacle and the corresponding displacement correction amount each time; when an additional carpet is added, the data recording module records the autonomous mobile device on the carpet The number of displacement corrections and the corresponding amount of displacement correction each time.

The data analysis module is connected to the data recording module, and analyzes two parts of the movement information recorded by the data recording module: the first part is when the new obstacle exists, the data analysis module analyzes the autonomous movement The device bypasses the movement information of the new obstacle and judges whether there is optimal movement information, that is, the new obstacle can be bypassed with the least number of displacement corrections; the second part is when the road friction coefficient changes, the data analysis model The group analyzes the movement information of the autonomous mobile device on the road surface to determine whether the number of times of displacement correction is greater than a threshold value. The so-called "threshold value" refers to a critical value set artificially according to actual conditions.

The data correction module is respectively connected to the data storage module and the data analysis module, and the data correction module is used to correct the map and map description file stored in the data storage module. When there is an obstacle, the data correction module corrects the map and the corresponding map description file; when only the road surface friction coefficient changes, the data correction module only needs to correct the map description file.

Referring to FIG. 2, the present invention further provides a method for automatically correcting environmental information, which includes the following steps:

S1, providing an autonomous mobile device that automatically corrects environmental information;

S2. Start the autonomous mobile device to move between two artificial markers, and the image acquisition module collects an image in front of its movement and transmits it to the artificial recognition module;

S3. The artificial mark recognition module recognizes the artificial mark in the image, and the autonomous mobile device determines its position relative to the artificial mark;

S4. The data recording module simultaneously records the change information and autonomy of the pavement environment between manual markings.

Movement information of a mobile device moving between the artificial marks, and the data analysis module determines optimal movement information to move between the artificial marks based on the recorded movement information;

S5. The data correction module corrects a map and a map description file in the data storage module according to the road surface environment change information and optimal movement information.

In step S3, the image acquisition module outputs the image containing the artificial mark to the artificial mark recognition module, and the artificial mark recognition module recognizes the ID of the artificial mark, thereby determining that the autonomous mobile device is relatively Manually mark the distance and angle so that the autonomous mobile device can autonomously correct its movement information, such as motor output and output time.

In step S4, the optimal mobile information refers to a next manual mark that can be reached with relatively few displacement correction times. The step of determining the optimal mobile information by the data analysis module based on the recorded mobile information is further include:

S41. When the road surface environment includes a new obstacle, record the number of displacement corrections of the autonomous mobile device to bypass the new obstacle, and the data analysis module analyzes whether there is a relatively small number of displacement corrections that can bypass the obstacle. For new obstacles, if "Yes", the movement information corresponding to the relatively small number of displacement corrections is optimal movement information; if "No", the movement information corresponding to the recorded displacement correction times is optimal movement information ;

S42. When the road surface environment is a change in road surface friction coefficient, record the number of displacement corrections and the amount of displacement correction when the autonomous mobile device moves on the road surface, and the data analysis module analyzes whether the number of displacement corrections is greater than a threshold value If "Yes", the data analysis module calculates the optimal movement information according to the number of displacement correction times and the amount of displacement correction; if "No", the movement information corresponding to the recorded displacement correction times is the optimal movement information .

In step S42, in this embodiment, the data analysis module obtains optimal movement information by averaging values according to the number of times of displacement correction and the amount of displacement correction.

In step S5, when the road surface environment is the addition or disappearance of an obstacle, the new obstacle is added or deleted from the pre-stored map, and the pre-stored movement information in the map description file is corrected to the optimal movement. Information; when the road surface environment is a change in road friction coefficient, the mobile information pre-stored in the map description file is corrected to the optimal mobile information.

The map stored in the autonomous mobile device and the corresponding map description file can be continuously modified by continuously performing steps S4 to S5, so that the smoothness of movement of the autonomous mobile device in the mobile area is gradually optimized.

The autonomous mobile device and method for automatically correcting environmental information according to the present invention, by storing a map of a region to be moved and a corresponding map description file in the autonomous mobile device, recording changes in road surface environment and corresponding mobile information during the movement between manual markers, The optimal mobile information is analyzed and determined, and the map and map description file stored in the autonomous mobile device are continuously and repeatedly revised based on the analysis, so that the autonomous mobile device moves more smoothly in the area to be moved.

Examples

As shown in the following figure, suppose a robot moves in a certain area of the factory, where artificial marks A and B are set in the area, so that the robot moves from the artificial marks A to B. A new obstacle F is set in this area. When moving to point c according to the pre-stored information, it was found that it could not continue to move forward. The robot autonomously moved 1 circle to the left to point d. However, it was found that it could not move forward along the original route, so it continued to move 3 circles to the left to reach At point e, it is found that you can move forward according to the original route, and move autonomously 4 times to reach point g. At this time, move to the right autonomously 4 times to reach point h, and move forward to the artificial mark B according to the original route. In the above process, the displacement was corrected a total of 4 times. Through analysis, it was found that the displacement can be corrected only once from point c to point e. The movement information in the file makes it necessary to correct the displacement only 3 times for the next movement. Through multiple movements, it is found that there are fewer movement modes than 3 times, and the original map description file can be continuously revised.

In summary, the present invention has indeed met the requirements for an invention patent, and a patent application was filed in accordance with the law. However, the above is only a preferred embodiment of the present invention, and it cannot be used to limit the scope of patent application in this case. For example, those who are familiar with the skills of this case and equivalent modifications or changes made in accordance with the spirit of the present invention should be covered by the following patent applications.

no

no

Claims (10)

An autonomous mobile device that automatically corrects environmental information includes:
A data storage module for storing a map of a region to be moved and a corresponding map description file in advance, a plurality of manual markers are set in the region to be moved, and an autonomous mobile device moves between the manual markers;
An image acquisition module, used to capture the front image of the autonomous mobile device during the movement;
An artificial mark recognition module for receiving an image signal output by the image acquisition module, and recognizing the artificial mark in the image, thereby realizing the positioning of the autonomous mobile device;
A data recording module, which records information about changes in the pavement environment between the artificial marks and movement information of autonomous mobile devices moving between the artificial marks;
A data analysis module, which analyzes the optimal movement information of the autonomous mobile device moving between the artificial marks according to the movement information recorded by the data recording module;
A data correction module for receiving signals of road surface environment change information recorded by the data recording module and signals of optimal movement information analyzed by the data analysis module, thereby correcting maps and map descriptions in the data storage module files. The autonomous mobile device for automatically correcting environmental information according to item 1 of the claim, wherein the manual tag is any one of a Tag36h11 tag series, a Tag36h10 tag series, a Tag25h9 tag series, or a Tag16h5 tag series. The autonomous mobile device for automatically correcting environmental information according to item 1 of the claim, wherein the map description file is used to store the relative distance, direction, and required motor output between any two artificial markers. The autonomous mobile device for automatically correcting environmental information according to item 1, wherein the mobile information includes a number of displacement corrections and a displacement correction amount when moving between the two artificial markers. The autonomous mobile device for automatically correcting environmental information according to item 1 of the claim, wherein the change information of the road surface environment includes changes in obstacles and changes in road friction coefficient. A method for automatically correcting environmental information includes the following steps:
Provide an autonomous mobile device that automatically corrects environmental information as described in claims 1 to 5;
Start the autonomous mobile device to move between two artificial markers, and the image acquisition module collects the image in front of the mobile and transmits it to the artificial recognition module;
The artificial mark recognition module recognizes the artificial mark in the image, and the autonomous mobile device determines its position relative to the artificial mark;
The data recording module simultaneously records the change information of the pavement environment between the manual marks and the mobile information of the autonomous mobile device moving between the manual marks. The data analysis module judges that the manual mark is based on the recorded movement information. Optimal mobile information between moves;
The data correction module corrects a map and a map description file in the data storage module according to the road surface environment change information and optimal movement information. The method for automatically correcting environmental information according to claim 6, wherein the optimal mobile information refers to an autonomous mobile device moving between two manual markers with a relatively small number of displacement corrections. The method for automatically correcting environmental information according to item 6, wherein the data analysis module determines the optimal mobile information based on the recorded mobile information further includes the following steps:
When the road surface environment includes a new obstacle, record the number of displacement corrections of the autonomous mobile device to bypass the new obstacle, and the data analysis module analyzes whether there are relatively few displacement correction times that can bypass the new obstacle. If it is "Yes", the movement information corresponding to the relatively small number of displacement correction times is the optimal movement information; if "No", the movement information corresponding to the recorded displacement correction times is the optimal movement information;
When the road surface environment is that the road friction coefficient becomes large, the number of displacement corrections and the amount of displacement correction when the autonomous mobile device moves on the road are recorded, and the data analysis module analyzes whether the number of displacement corrections is greater than a threshold value, If “Yes”, the data analysis module calculates the optimal movement information according to the number of times of displacement correction and the amount of displacement correction; if “No”, the movement information corresponding to the recorded number of displacement corrections is the optimal movement information. The method for automatically correcting environmental information according to item 8 of the claim, wherein the threshold value is a critical value set artificially according to actual needs. The method for automatically correcting environmental information according to item 8 of the claim, wherein the specific process for the data correction module to modify the map and map description file in the data storage module is: when the road surface environment is an addition of obstacles When it disappears, the new obstacle is added or deleted from the pre-stored map, and the pre-stored movement information in the map description file is corrected to the optimal movement information; when the road surface environment is the change of the road friction coefficient, the The pre-stored mobile information in the map description file is corrected to the optimal mobile information.